TLV9352: Why is the step response of is slow?

Part Number: TLV9352
Other Parts Discussed in Thread: OPA2991, OPA1688, OPA2197, OPA2182, OPA196, OPA191, OPA192

Tool/software:

Dear Specialists,

My customer is evaluating TLV9352, they are encountering the step response is slow.

The customer would like to know what could be the cause.

I would be grateful if you could advise.

---Situation

Circuit

Non inverting Gain 14 

Could you please see attached excel file?

Conditions

-Input: applied in steps of -0.71V to 0.71V.

-Output: assumed to change from -10V to 10V.

Results

-Input: stable at 1us

-Output: stable at 6.5us

-The behavior of the output voltage rise is close to the slew rate in the data sheet at first,
but then become at 0.5V/us halfway through.

Question about TLV9352.xlsx

---

I appreciate your great help in advance.

Best regards,

Shinichi

  • Hi Shinichi-san,

    The behavior of the output voltage rise is close to the slew rate in the data sheet at first,
    but then become at 0.5V/us halfway through.

    The output initial response seems to match what the datasheet is specified, where the output's rise time is approx. 1.2 usec ( based on 10% to 90% slew rate at the edge response in timing). 

    The TLV9352 may have boosted slew rate implementation in the IC, where the long tail response may be contributed and returned to natural slew rate in the design. I did not see that the feature is described in the datasheet, and I have to verify this. In other words, ~0.5V/usec may be returned to its natural slew rate when the output is approaching to its setpoint. 

    https://www.ti.com/lit/an/sloa332a/sloa332a.pdf?ts=1739847237513

    I do not know what is your design requirements in the output response timing. If the overall step response is not meeting the customer's requirements, we may have to find alternative op amp to meet the requirements. Please specify the cost and performance requirements in the application. 

    Best,

    Raymond

  • Hi Raymond,

    Thank you for your reply.

    When I looked at the waveform, I also thought it might be the effect of the slew boost.

    However, this was not mentioned in the data sheet.

    Please investigate this as soon as possible.

    If the slew boost is not equipped, could you please look into the cause?

    I appreciate your great help and cooperation.

    Best regards,

    Shinichi

  • Hi Shinichi,

    When I looked at the waveform, I also thought it might be the effect of the slew boost.

    Yes, TLV9352 (GP op amp) has slew boost feature. Typically, the natural slew rate may be 1/10 of the boosted 20V/usec. I do not know how the designer is implemented in the Silicon, and I have to locate the designer for these figure. 

    This was designed by GPAMPS and it may take a few days to get the reply back from the designer. I will let you know. 

    Best,

    Raymond  

  • Hi Shinichi,

    TLV9352 design is based on OPA2991 core, where the slew rate is about 0.5V/usec when Vid is input below 100mV, see the plot below. This is in agreement from what you observed in your test results. 

    If you have additional questions, please let us know. 

    Best,

    Raymond

  • Hi Raymond,

    Thank you for your reply.

    I understand TLV9352 has a function of slew boost.

    I'll share this information with the customer.

    As a substitute, I selected OPA2197 and OPA1688.

    If you have the other candidate, could you please let me know.

    I appreciate your great help and cooperation.

    Best regards,

    Shinichi  

  • Hi Raymond,

    OPA2197 has a slew boost.

    Therefore, OPA1688 is the only candidate.

    Shinichi

  • Hi Shinichi, 

    TLV9352 belongs to general purpose op amp family and OPA2197 belongs to Precision op amp family.

    OPA2197 has a slew boost.

    Yes, however, OPA2197 should not see the long tail response as in TLV9352. What is your step response requirement for the application? 

    OPA2197 will have a slight delay, but it is around 1.8usec or better, not 6.5usec. 

    OPA2197 Step Response timing 02212025.TSC

    In precision op amp family, we have numbers of op amps to choose from, see the attached list. There are other high slew rate op amp among the options. 

    https://www.ti.com/amplifier-circuit/op-amps/precision/products.html?login-check=true#480=2&23typ=10%3B20&sort=22typ;desc&

    If you need further assistant, please let me know. 

    Best,

    Raymond

  • Hi Raymond,

    Thank you for your reply.

    I understand OPA2197 may be a substitute.

    The other device, I'll confirm again.

    By the way, the customer has an additional question.

    The customer cannot confirm the slew boost operation with the PSpice mode simulation..

    Does TLV9352 PSpice simulation model include Slew Boost function?

    I appreciate your great help and cooperation.

    Best regards,

    Shinichi

  • Hi Raymond,

    The customer is concerned.
    Is it possible that OPA2197 is not reflected in the simulation model, just like TLV9352?

    Could you please measure step response using an actual OPA2182?

    I appreciate your great help and cooperation.

    Best regards,

    Shinichi

  • Hi Shinichi, 

    Is it possible that OPA2197 is not reflected in the simulation model, just like TLV9352?

    I have replied previously that TLV9352 is originated from general purpose op amp team, and opa2197 is a part of precision op amp team (PRAMPS) Two teams are two different business entities in TI. 

    Being in precision op amp application team in supporting our products, I never encountered issues with OPA2197 or OPA219x (OPA192, OPA196, OPA191 etc.) family about the step response delay due to slew boost feature in PRAMPS.  The OPAx19x family is used for Mux friendly op amps in interfacing with ADC and its related application, see the figure below. And its step response behavior is totally different from some of the general purpose op amp (e.g. TLV935x or TLV915x families). 

    There will be a few usec delay, but it is nothing close to TLV9352's step response behaviors. 

    Could you please measure step response using an actual OPA2182?

    I can do the measurement on your behalf, but you have to tell me the step response and driving load requirements. In addition, I may have to order OPA2197 and OPA2182 op amps from TI E-store (I do not have these in the lab, and I do not know our validation team has these in stock either). Anyway, it will take time to receive the op amp in the lab and tested for you.

    Please provide the step response requirements including the op amp package information that you are interested, I can perform these tests. Please keep in mind that it may take up to 10 business days to get the test results. 

    If you want to proceed with the request, please let me know. 

    Best,

    Raymond 

  • Hi Raymond,

    Thank you for your reply.

    As you mentioned, OPA2197 is able to confirm the step response.

    I'll share this information with the customer.

    Regarding TLV9352, it is not described about slew boost in the datasheet.

    I think it is required to add a  slew boost function description and step response waveform in the datasheet.

    Could you please consider?

    I appreciate your great help and cooperation.

    Best regards,

    Shinichi 

  • Hi Shinichi, 

    As you mentioned, OPA2197 is able to confirm the step response.

    Great. you tested and figured it out. Yes, our the precision op amps do not have the slew rate issues for small input signals, even though it has the slew boost feature. 

    I think it is required to add a  slew boost function description and step response waveform in the datasheet.

    I have suggested to revise the slew boost  description in the datasheet for these general purpose op amps - this is done by our system engineer team. Unfortunately, the datasheet revision is a long and time consuming process; it has multiple approval process at various level.

    I am going to close this query. Thanks for your feedback!

    Best,

    Raymond

  • Hi Raymond,

    The customer understand the situation.

    I agree to close this thread.

    I appreciate your great help and cooperation.

    Best regards,

    Shinichi